Weird Computer issue.

A

AndrewClaycomb

New member
My computer is acting really weird when I turn off my Firepod. It hasn't always done this though.


I've had the Firepod for a little more than a week now (I think), and I keep it on a second power strip with my powered monitors. The reasoning is I can turn them all on/off with the flip of a single switch, which is nice when I'm not using the audio stuff.

Today I got my firewire external hard drive which turns on and off with my computer. The Firepod is slaving off the hard drive. (I only have 1 Texas Instruments Firewire port.) Everything works great when the Firepod is on but if I turn it off ,my computer will slowly start locking up. Applications stop responding until I just turn off my computer manually. Task Manager won't work when this happens either.

Then when I turn the computer back on (with the Firepod off) Windows won't load.

If the Firepod is on, Windows will load and run completely fine.

WTF?

EDIT: To add to the weirdness, if I don't have either the hard drive or the Firepod on, it works fine. My temporary fix is to keep the hard drive on the same power strip as the Firepod. That way they both will only turn on at the same time. I only use this new drive for recording anyway, but it would be nice to use it all the time for all my storage.
 
Some device have weird FireWire bugs that only manifest themselves when the devices are off. In theory, bus power is supposed to be enough to keep the PHY (the physical link layer hardware) running, but if you have devices in a chain, there may not be enough bus power to reliably keep the FIREPOD's PHY powered up and in a power-starved state, its PHY may be generating truckloads of bus resets or something weird.

Have you checked your system logs or whatever to see if you're seeing truckloads of bus resets? (I assume Windows has some sort of system logging.... Haven't touched the stuff in years.)

You might try switching the order and putting the FIREPOD first except that I have a vague memory that perhaps the FIREPOD doesn't pass traffic when turned off... or am I thinking of the M-Audio FW1814? I forget....
 
it seems to work reliable when I reverse the order, however it won't pass signal as you said. So I'm still stuck with only being able to use the hard drive with the Firepod on. It's not a huge deal, because I record directly to the drive and that's basically it (aside from back ups)

It would have been nice to use it all the time.

I have no idea how to check logs on XP.
 
Perversely, you might try using a 6-4 converter and a 4-6 converter to ensure that no bus power reaches the FIREPOD so that its FireWire chips don't see any power except when it is turned on. If the problem is caused by a broken bus power circuit design in the FIREPOD as I suspect, that will fix it. If that doesn't fix it, then I don't know what's wrong....
 
AndrewClaycomb said:
I didn't mention that before, I'm running everything off a single 4 pin port. So I'm not sending any power from my computer.
Click to expand...

Ah. That means that any bus power that the FIREPOD's PHY is running off of would be coming from your drive. Could be that the drive's power supply is a little too dirty for the FIREPOD's power filtering... if the FIREPOD's even tries to power its PHY silicon off bus power....

Oh, another question. If you boot up with the FIREPOD turned off, do you still have problems, or only if you turn it off after the HD is turned on? I ask because it is possible the hard drive's PHY silicon is brain damaged and isn't handling device disconnection correctly (and/or that Windows' support for that particular hub silicon is brain damaged and isn't handling device disconnection correctly).

I'm really grasping at straws here, though. That's really odd behavior. :)
 
far too many variables to add anything particularly helpful

(chip set of firewire card etc.)

locating and interpreting the event logs might help (well might help someone) administrative tools in control panel will get you to MS defaults thishttp://support.microsoft.com/kb/308427

might help you sort out some of the log issues

& it is a place to start if you want diagnosis

the 'firewire reset storm', particularly while diasy chaining is pretty common . . . but there is no one size fits all fix

firewire cards are pretty cheap your best bet might be to un daisy chain

additionally presence of a firewire network adaptor (which it is unlikely you need) can cause issues

specific hardware responds differently on different systems

have two nearly identical Thinkpad A31 (laptops) using the same cardbus firewire interface an Onyx firewire card has to be plugged in after system boots on one and can be plugged in during boot on the other

event logs can provide some idea of what is happening and suggest possible other routes to investigate . . . without that info everything else is random guessing

my guess is undaisy chain
 
dgatwood said:
Ah. That means that any bus power that the FIREPOD's PHY is running off of would be coming from your drive. Could be that the drive's power supply is a little too dirty for the FIREPOD's power filtering... if the FIREPOD's even tries to power its PHY silicon off bus power....
Click to expand...

This makes a lot of sense. Both have their own power supplies, so I was assuming they didn't need to pass power at all. But if they are, I could see how dirty power coming from the drive could wreak havok. My issue with that is, when the Firepod is off shouldn't everything then work fine? In this order, the only time my system runs correctly is when they are both on. And I'm not getting pops or clicks through my interface.

Since this happened I've reversed the order and everything works fine, I just can't have my drive on unless I'm working with audio.


dgatwood said:
Oh, another question. If you boot up with the FIREPOD turned off, do you still have problems, or only if you turn it off after the HD is turned on? I ask because it is possible the hard drive's PHY silicon is brain damaged and isn't handling device disconnection correctly (and/or that Windows' support for that particular hub silicon is brain damaged and isn't handling device disconnection correctly).

I'm really grasping at straws here, though. That's really odd behavior. :)
Click to expand...

With the original chaining order (Computer->Hard Drive->Firepod), If I tried to boot the computer with the Firepod off and the hard drive on (Turns on by itself when computer boots), Windows will not load. It goes through the image of "Windows" with the loading bar beneath and it doesn't load.
 
oretez said:
far too many variables to add anything particularly helpful

(chip set of firewire card etc.)

locating and interpreting the event logs might help (well might help someone) administrative tools in control panel will get you to MS defaults thishttp://support.microsoft.com/kb/308427

might help you sort out some of the log issues

& it is a place to start if you want diagnosis

the 'firewire reset storm', particularly while diasy chaining is pretty common . . . but there is no one size fits all fix

firewire cards are pretty cheap your best bet might be to un daisy chain

additionally presence of a firewire network adaptor (which it is unlikely you need) can cause issues

specific hardware responds differently on different systems

have two nearly identical Thinkpad A31 (laptops) using the same cardbus firewire interface an Onyx firewire card has to be plugged in after system boots on one and can be plugged in during boot on the other

event logs can provide some idea of what is happening and suggest possible other routes to investigate . . . without that info everything else is random guessing

my guess is undaisy chain
Click to expand...

I think getting a pcmcia firewire card with 2 ports will be my next course of action. I'll go through the logs and see what I can find, thanks for posting that information.
 
AndrewClaycomb said:
This makes a lot of sense. Both have their own power supplies, so I was assuming they didn't need to pass power at all. But if they are, I could see how dirty power coming from the drive could wreak havok. My issue with that is, when the Firepod is off shouldn't everything then work fine?
Click to expand...

When you're dealing with a FireWire device, there are two aspects to the operation of the device. You have some sort of controller silicon (in this case, BeBoB) and some sort of PHY silicon (usually TI). That PHY silicon, among other things, handles the FireWire communication.

The PHY silicon also acts like a hub between the ports, however, and for this reason, the standard way to wire them is to use the bus power as a secondary power supply for the PHY so that you can completely shut off the device's power supply and have the hub functionality still work (still pass traffic from one port to the other).

When you shut off power to the FIREPOD, it is no longer powering the PHY. How the PHY behaves in that situation depends on whether it is wired normally (to run off of bus power if available) and on whether the other hardware in the chain is providing enough bus power to keep it running reliably or not. :)


AndrewClaycomb said:
With the original chaining order (Computer->Hard Drive->Firepod), If I tried to boot the computer with the Firepod off and the hard drive on (Turns on by itself when computer boots), Windows will not load. It goes through the image of "Windows" with the loading bar beneath and it doesn't load.
Click to expand...

Okay, that eliminates all the easy causes. No matter how hosed the FireWire bus is, it shouldn't keep Windows from booting, I wouldn't think, though I guess anything is possible. I suppose it is also remotely possible that whatever the FIREPOD is doing is making the controller silicon in the HD so unhappy that it is DMAing random crap all over memory.... Are there any firmware updates available for that hard drive, by any chance? You might try disabling FireWire DMA support (again, though, no idea how to do that in XP) and see if that helps.
 
You must log in or register to reply here.
Back
Top